The following disclosure is based on German Patent Application No. 10 2018 216 870.9 filed on October 1 , 2018, which is incorporated into this application by reference.

FIELD OF APPLICATION AND PRIOR ART

The invention relates to a method for producing an illumination system for an EUV apparatus in accordance with the preamble of claim 1 and to an illumination system for an EUV apparatus in accordance with the preamble of claim 9. The EUV apparatus can be, for example, a projection exposure apparatus for EUV microlithography or a mask inspection apparatus, employing EUV radiation, for inspecting masks (reticles) for EUV microlithography.

Lithographic projection exposure methods are predominantly used nowadays for producing semiconductor components and other finely structured components, such as, for example, masks for photolithography. In this case, use is made of masks (reticles) or other patterning devices that bear or form the pattern of a structure to be imaged, for example a line pattern of a layer of a semiconductor component. The pattern is positioned in a projection exposure apparatus between an illumination system and a projection lens in the region of the object plane of the projection lens and is illuminated with an illumination radiation shaped by the illumination system. The radiation modified by the pattern travels through the projection lens as projection radiation, said projection lens imaging the pattern with a reduced scale onto the substrate to be exposed. The surface of the substrate is arranged in the image plane of the projection lens, which image plane is optically conjugate to the object plane. The substrate is generally coated with a radiation-sensitive layer (resist, photoresist).

One of the aims in the development of projection exposure apparatuses is to lithographically produce structures having smaller and smaller dimensions on the substrate, for example to obtain greater integration densities in semiconductor components. One approach consists in working with shorter wavelengths of the electromagnetic radiation. By way of example, optical systems have been developed which use electromagnetic radiation from the extreme ultraviolet range (EUV), in particular having operating wavelengths in the range of between 5 nanometres (nm) and 30 nm, in particular of 13.5 nm.

An EUV projection exposure apparatus comprising an illumination system is known e.g. from patent US 7 473 907 B2. The illumination system is embodied for receiving EUV radiation of an EUV radiation source and for shaping illumination radiation from at least one portion of the received EUV radiation. The illumination radiation is directed into an illumination field in an exit plane of the illumination system during exposure operation, wherein the exit plane of the illumination system and the object plane of the projection lens advantageously coincide. The illumination radiation is characterized by specific illumination parameters and is incident on the pattern within the illumination field with a defined position, shape and size at defined angles. The EUV radiation source, which may be a plasma source, for example, is arranged in a source module separate from the illumination system, said source module generating a secondary radiation source at a source position in an entrance plane of the illumination system.

Arranged in a housing of an illumination system of the type considered here are a plurality of mirror modules, which are each located in the final installed state at installation positions that are provided for the mirror modules. The mirror modules or reflective mirror surfaces of the mirror modules define an illumination beam path extending from the source position to the illumination field.

DE 10 2016 203 990 A1 describes methods for producing an illumination system of this type, wherein production comprises an adjustment. The method involves installing mirror modules of the illumination system at installation positions provided for the mirror modules to establish an illumination beam path which extends from the source position to the illumination field, Measurement light is coupled into the illumination beam path at an input coupling position upstream of a first mirror module of the illumination beam path and is detected after reflection at each of the mirror modules of the illumination beam path by a detector. Actual measurement values for at least one system measurement variable are ascertained from the detected measurement light, wherein the actual measurement values represent an actual state of the system measurement variable of the illumination system. Correction values are ascertained from the actual measurement values, said correction values being used for the adjustment. This involves adjusting at least one mirror module with variation of the orientation thereof in the installation position in degrees of freedom of a rigid body using the correction values to change the actual state in a manner such that in the case of irradiation with EUV radiation from the EUV radiation source, the illumination radiation in the illumination field fulfils the illumination specification. Measurement light from the visible (VIS) spectral range is preferably used in this case.

DE 10 2012 010 093 A1 describes an EUV illumination system comprising a facet mirror having a multiplicity of facets, each having an IR diffraction structure designed such that at least one portion of incident radiation from the infrared range is diffracted out of the illumination beam path. That wavelength of the IR radiation which is to be diffracted out is, in particular, the wavelength of a laser used for generating a plasma in the EUV radiation source.

PROBLEM AND SOLUTION

One problem addressed by the invention is that of specifying a method for producing an illumination system which allows reliable production of well-adjusted illumination systems of the type mentioned in the introduction. A further problem addressed is that of providing a well- adjusted illumination system.

To solve this problem, the invention provides a method for producing an illumination system having the features of Claim 1 , an illumination system having the features of Claim 9, and a measurement light source module having the features of Claim 15. Advantageous developments are specified in the dependent claims. The wording of all the claims is incorporated by reference in the content of the description.

The method involves installing mirror modules of the illumination system at installation positions provided for the mirror modules to establish an illumination beam path which extends from the source position to the illumination field. In the context of an adjustment, measurement light is coupled into the illumination beam path at an input coupling position upstream of a first mirror module of the illumination beam path and is detected after reflection of the measurement light at each of the mirror modules of the illumination beam path. Actual measurement values for at least one system measurement variable are ascertained from the detected measurement light, wherein the actual measurement values represent an actual state of the system measurement variable of the illumination system. Correction values are ascertained from the actual measurement values. The method involves adjusting at least one mirror module using the correction values to change the actual state in a manner such that in the case of irradiation with EUV radiation from the EUV radiation source, the illumination radiation in the illumination field fulfils the illumination specification. The method can be used during the original production of the illumination system by the manufacturer in the context of an initial adjustment. The wording “producing an illumination system” also encompasses use during reestablishment, e.g. after exchange of a mirror module, in particular by the end user at the location of the earlier use.

By way of example, using sensitivities that represent a relationship between the respective system measurement variable and a change in the orientation of at least one mirror module in its installation position, correction values can be ascertained from the actual measurement values. The correction values indicate how the actual state must be changed in order to approach the desired state. On the basis of the measurement, the correction values can then be used to carry out an adjustment of at least one mirror module with variation of the orientation of the mirror module in its installation position, i.e. in its degrees of freedom of a rigid body, in order to change the (measured) actual state in a manner such that, during use as intended, in the case of irradiation with EUV radiation from the EUV radiation source, the illumination radiation in the illumination field fulfils the illumination specification. By way of example, the position of the illumination field in space can be changed such that it corresponds as well as possible to the position of the object field of the downstream imaging optical unit.

The method is used to produce an illumination system wherein at least one of the mirror modules has a mirror surface on which an IR diffraction structure is embodied, which is designed such that at least one portion of incident radiation from the infrared range (IR radiation) is diffracted out of the illumination beam path. What can be achieved is that those portions of infrared radiation which are diffracted out of the illumination beam path cannot pass or cannot directly pass into the illumination field in the exit plane of the illumination system and into downstream systems. Radiation from the infrared range comprises, in particular, electromagnetic radiation in the wavelength range of 780 nm to 1 mm. IR radiation can lead e.g. to heat-induced disturbances in projection exposure apparatuses (e.g. heating of optical components, imaging aberrations caused as a result).

By way of example, the IR diffraction structure can be designed such that infrared radiation from the infrared range, for example around approximately 10.6 pm, is diffracted out of the illumination beam path. Radiation having such IR portions can pass into the illumination beam path e.g. if a primary EUV light source in the form of an LPP (laser-produced plasma) source is used, in which tin is excited to form a plasma by means of a carbon dioxide laser operating at a wavelength of 10.6 pm. Other target wavelength ranges are also possible. By way of example, it is possible to optimize the diffraction efficiency for IR wavelengths around 1 pm, e.g. in cases in which an Nd:YAG laser which emits at 1064 nm is used in the EUV radiation source.

In the method, measurement light having a wavelength l from the visible spectral range (VIS) is used for adjustment purposes. This has proved to be particularly advantageous with regard to handling and outlay during the measurements and allows measurements with high measurement accuracy. Within the meaning of this application, the visible spectral range includes, in particular, wavelengths in the range of 380 nm (violet) to 640 nm (red).

According to the claimed invention, the wavelength of the measurement light is selected in such a way that higher orders of measurement light diffracted at the IR diffraction structure are substantially suppressed. To put it another way, that means that a predominant portion of the measurement light intensity is diffracted into the 0 order (zero order) at the IR diffraction structure. The inventors have discovered that measurement light of this type does not impair the measurement or does not impair it in any relevant way. It has thus been recognized that even in illumination systems in which at least one mirror surface bears an IR diffraction structure of the type mentioned, precise measurements with wavelengths from the visible wavelength range can be carried out if the correct wavelength (measurement wavelength) is selected.

The IR diffraction structure can be designed such that it has no significant diffractive effect for the EUV radiation used during operation of the illumination system. This should be understood to mean, in particular, that the intensity of the zero order of diffraction of the EUV radiation that is required for the desired imaging is reduced by at most 10%, in particular by at most 5%, preferably by at most 1% or less, for example a maximum of 0.1%, by the IR diffraction structure.

The inventors have discovered that, in contrast thereto, the IR diffraction structures can be a disturbance during the measurements with measurement light from the visible spectral range. In the case of typical structure dimensions suitable for diffracting IR radiation, the diffraction angles for visible light can be so small that in the illumination field the ± 1st orders (i.e. the - 1 st and the + 1 st order of diffraction) are so close to the 0 order of diffraction that they still fall into the illumination field. The measurements can be disturbed as a result. By way of example, the measurements for determining the position of the illumination field can be disturbed.

This disturbing effect can be largely or completely avoided upon the implementation of the invention, i.e. upon the selection of suitable wavelengths for the measurement, by the wavelength of the visible light being selected such that higher orders of diffraction (first and higher orders) arising at the IR diffraction structure are largely suppressed. Suppression of higher orders of diffraction is present in particular if the intensity at higher orders of diffraction is at most 10% of the intensity at the zero order of diffraction, in particular at most 5%. To put it another way, in the context of this application, higher orders of diffraction are deemed to be “substantially suppressed” in particular if the intensity at a higher order of diffraction is at most 10%, preferably at most 5%, in particular at most 2%, of that intensity which is present at the 0 order of diffraction.

An IR diffraction structure can be aperiodic (non-periodic) or periodic or have both aperiodic and periodic portions or regions. The IR diffraction structure is preferably designed as a diffraction grating, that is to say as a periodic structure that can be characterized by a grating depth d and a grating period p. In the case of a reflective phase grating in the form of an echelon grating, the grating depth corresponds to the step height. Given a grating depth d, the wavelength l of the measurement light is preferably selected in such a way that the grating depth d substantially corresponds to a half-integral multiple of the wavelength of the measurement light, such that the condition (d/cos(a) = h·l/2) is substantially satisfied. In this case, a is the angle of incidence of the measurement light with respect to the normal upon incidence on the IR diffraction structure, and n is a positive integer. The wavelength l of the measurement light is thus preferably chosen such that the best possible constructive interference arises in the measurement light reflected by the diffraction grating.

When mention is made of a“wavelength of the measurement light” in this application, this does not involve a single wavelength, but rather a wavelength range or a wavelength band of adjacent individual wavelengths that has a greater or lesser spectral bandwidth.

In one development, a particularly good adaptation of the wavelength of the measurement light to the incidence situation at the IR diffraction structure is achieved by virtue of the fact that during a measurement with a selected wavelength of the measurement light, a spectral bandwidth of the measurement light is less than 2 nm (Full Width Half Maximum = FWHM). The measurement light should thus be sufficiently narrowband such that, for a given structure of the IR diffraction structure, the greatest possible proportion of the intensity of the measurement light can fall into the zero order of diffraction.

The abovementioned condition (d/cos(a) = h·l/2) is substantially satisfied in particular if the measurement light has a bandwidth of at most 2 nm, and the condition is exactly satisfied for a wavelength within this bandwidth. To put it another way, it should be taken into account that the abovementioned condition can only ever be exactly satisfied for one wavelength. What is crucial for this partial aspect is that suppression of the higher orders also still takes place if a sufficiently small range (2nm) around the exact wavelength is employed.

An EUV illumination system is generally designed such that during operation the EUV radiation can be radiated into the illumination field from different directions. In general, the intention is to be able to provide different illumination settings, that is to say different illumination intensity distributions in a Fourier plane of the field plane, the so-called pupil plane of the illumination system. In illumination systems comprising a field facet mirror and a pupil facet mirror, different illumination channels are generally provided for the illumination, wherein an illumination channel is a partial beam path extending from the source position to the illumination field via a field facet and an assigned pupil facet and, if appropriate, further mirror surfaces. Depending on the location of an IR diffraction structure in the illumination beam path, it may be the case that measurement light from different illumination channels is incident on the same IR diffraction structure from different angles of incidence. In order in this case to achieve sufficiently great suppression of higher orders of diffraction with the use of measurement light from the visible wavelength range, it can be advantageous if during the input coupling of measurement light, different beam angles of the coupled-in measurement light into the illumination beam path are set, wherein different wavelengths of the measurement light are used for individual illumination channels or subgroups of illumination channels with similar beam angles. To put it another way, the different beam angles can be taken into account in the beam path by virtue of different wavelengths of the measurement light being used for individual illumination channels or subgroups of illumination channels with similar beam angles. In this case, by way of example, measurement light that can be used for illumination channels whose angle of incidence on the IR diffraction structure is particularly large can have a longer wavelength than measurement light used for illumination channels impinging with normal or almost normal incidence on the IR diffraction structure.

The inventors have ascertained that it can be advantageous if during the measurement with different wavelengths of the measurement light, a spectral range of different wavelengths of the measurement light having a bandwidth of at least 20 nm is used, wherein said bandwidth can, if appropriate, also be up to 30 nm or more. What can be achieved under these conditions in general is that, for all angles of incidence that are possible within the illumination system, it is possible to achieve the sought suppression of the higher orders of diffraction of the measurement light to a sufficiently great extent.

There are various possibilities for providing the measurement light.

In some embodiments, a measurement light source module that is tunable continuously variably or in steps is used for generating the measurement light. The measurement light source module can have a tunable primary measurement light source that is able to emit different wavelengths depending on the setting. This can be a tunable laser light source, for example.

It is also possible for the measurement light source module to have a broadband primary measurement light source and a downstream adjustable device for wavelength selection. The measurement light can thus be emitted by a broadband primary measurement light source, e.g. an LED or a thermal radiator, and be subjected to wavelength selection before being coupled into the illumination beam path, in order to obtain the desired measurement wavelength (i.e. a narrower wavelength range). In one embodiment, provision is made of a measurement light source module having a primary measurement light source in an entrance plane, and a 4f imaging system for imaging the primary measurement light source into a secondary measurement light source is disposed downstream of said primary measurement light source in an exit plane that is conjugate to the entrance plane, wherein a tiltable interference filter is arranged in a Fourier plane between the entrance plane and the exit plane. Said filter is situated in the parallel or almost parallel beam path between entrance plane and exit plane and can be designed such that, depending on the tilt angle from the arriving broadband light only a narrow band range is allowed to pass or is transmitted as measurement light. The interference filter can be tilted in order to tune the wavelength continuously variably in a certain range. The preferably continuously variably tiltable interference filter can be arranged in an interchangeable holder, such that the interference filter can be exchanged for another interference filter of the same type or of a different type without demounting of the device. This can be advantageous, for example, if a single interference filter cannot cover the entire required wavelength range of the measurement light. For this case, a plurality of exchangeable interference filters can be provided, which can be moved into the beam path of the measurement light source module if appropriate in an automated manner by means of a turret arrangement or a linear stage.

A measurement light source module can be mounted directly at suitable interface structures of the illumination system for the coupling of the measurement light source module. Other mounting possibilities can be provided as necessary. In accordance with one development, the measurement light is guided from an exit plane of the measurement light source module to the source position via a, preferably flexible, optical waveguide. As a result, it is possible to mount the measurement light source module with its optical and other components at a spatial distance from the illumination system at a suitable location. The exit of the optical waveguide then generally serves as a measurement light source at the illumination system. The term “optical waveguide” denotes here, in particular, transparent components such as fibres, tubes or rods which can transport light (here: measurement light) over short or long distances. In this case, the guiding of light is achieved by reflection at an interface of the optical waveguide either by total internal reflection on account of a lower refractive index of the medium surrounding the optical waveguide, or as a result of reflective coating of the interface. The optical waveguide can comprise an optical glass fibre or an optical glass fibre bundle, for example.

The aspect of using an optical waveguide for feeding measurement light into the illumination beam path and the spatial separation - possible as a result - between measurement light source module and illumination system during the measurement can also be advantageous during measurements at illumination systems having mirrors without IR diffraction structures. By way of example, the measurement system described in DE 10 2016 203 990 A1 can have an optical waveguide between the measurement light source module and the input coupling location at the illumination system.

Another aspect of the present invention relates to an exchangeable measurement light source module configured as a self-contained functional module which can be provided separately from the illumination system and which can be functionally connected to the illumination system or removed therefrom as needed. The measurement light source module is configures as a functional component of a measuring system, where other functional components of the measuring system are attached at and/or integrated into the structure of the illumination system. The measurement light source module can be configured to be capable of being mounted directly at suitable interface structures of the illumination system for the coupling of the measurement light source module. The measurement light source module may comprise complementary interface structures. Alternatively, measurement light source module may be configured to provide measurement light which may be guided from an exit plane of the measurement light source module to the source position of the illumination system via an optical waveguide, which may be rigid or flexible. In some embodiments, the measurement light source module is tunable continuously or in steps to generate measurement light at different wavelength ranges as needed. Alternatively it is possible to provide a set of two or more exchangeable measurement light source modules, each configured to provide measurement light from a single predefined wavelength (narrow wavelength range) only, where the predefined wavelength ranges differ between the different measurement light source modules of the set.

In many embodiments, the mirror modules comprise a first mirror module having a first facet mirror at a first installation position and a second mirror module having a second facet mirror at a second installation position of the illumination system. A mirror module of this type has a main body acting as a carrier, on which facet elements with reflective facets are mounted alone or in groups in accordance with a specific local distribution. If the reflective facets of the first facet mirror are situated at or near a field plane of the illumination system that is conjugate to the exit plane, the first facet mirror is frequently also referred to as a “field facet mirror.” Correspondingly, the second facet mirror is frequently also referred to as a“pupil facet mirror” if the reflective facets thereof are situated at or near a pupil plane which is Fourier-transformed in relation to the exit plane. The two facet mirrors contribute in the illumination system of the EUV apparatus to the homogenization or mixing of the EUV radiation.

In many method variants, with the aid of the measurement system, at least three system measurement variables or performance measurement variables are acquired, namely (i) the position of the illumination field at the reticle level or in the exit plane of the illumination system (corresponding to the object plane OS of the projection lens); (ii) the spatial distribution of measurement light in a pupil plane of the illumination system which is Fourier-transformed in relation to the exit plane, which determines the telecentricity at the reticle level or in the exit plane, and (iii) a luminous spot impingement on pupil facets, i.e. the position of a measurement light spot on a facet of the second facet mirror. For details, reference is made to DE 10 2016 203 990 A1 , the disclosure content of which in this respect is incorporated by reference in the content of this description.

The method can be used in the original production of the illumination system, i.e. during first production (first installation), to initially adjust the installed mirror modules with respect to their orientation such that the illumination system in the finished mounted state fulfils the illumination specification. A separate measuring machine containing all the components of the measurement system can be provided for this initial adjustment. These components generally include a measurement light source module, with which measurement light is produced, and a detector module, with which the measurement light, after it has passed through the relevant part of the illumination beam path (reflection at all mirror modules), is detected and prepared for evaluation. The measuring machine can contain a measurement frame, in which the frame of the illumination system can be installed. Using a positioning system, the frame of the illumination system can be moved into the correct position with respect to the measurement light source module and the detector module so that the measurement can be performed.

However, other cases are also possible, in which an illumination system of an EUV apparatus by an end user has already been in operation for a prolonged period of time and an adjustment should be performed as part of maintenance works. It is in particular also possible for a mirror module, after prolonged proper use under EUV irradiation, to change its properties so considerably due to optical, thermal and/or mechanical influences that it needs to be demounted and replaced by a different but not yet defunct mirror module of nominally identical or similar construction. This is also possible within the scope of the claimed invention. The wording “producing an illumination system” thus also encompasses reestablishment, in particular by the end user at the location of earlier use. Possible prerequisites and strategies for mirror module exchange are described in DE 10 2016 203 990 A1 , the disclosure content of which in this respect is incorporated by reference in the content of this description. BRIEF DESCRIPTION OF THE DRAWINGS

Further advantages and aspects of the invention are evident from the claims and from the following description of preferred exemplary embodiments of the invention, which are explained below with reference to the figures.

Fig. 1 shows optical components of an EUV microlithographic projection exposure apparatus in accordance with one embodiment of the invention;

Fig. 2 shows a number of ray trajectories in a mirror arrangement having two facet mirrors;

Figs. 3A and 3B show examples of the possible embodiment of IR diffraction structures in the form of binary phase gratings at a multilayer mirror;

Fig. 4 schematically shows components of an illumination system in accordance with another exemplary embodiment;

Fig. 5 shows a d-l diagram concerning the relationship between the step depth d of a binary phase grating and the wavelength l of the measurement light for various angles of incidence a of measurement light;

Fig. 6 shows components of a measurement light source module which enables a continuously variable setting of different measurement wavelengths in combination with a setting of different emission angles;

Fig. 7 shows components of a measurement light source module with a downstream optical waveguide for guiding measurement light to a remote location.

DETAILED DESCRIPTION OF THE EXEMPLARY EMBODIMENTS

Figure 1 shows by way of example optical components of an EUV microlithographic projection exposure apparatus WSC in accordance with one embodiment of the invention. The EUV microlithographic projection exposure apparatus serves during operation for exposing a radiation-sensitive substrate W, which is arranged in the region of an image plane IS of a projection lens PO, with at least one image of a pattern of a reflective mask (here also alternatively referred to as reticle), said pattern being arranged in the region of an object plane OS of the projection lens. In the case of the example, the substrate is a wafer composed of semiconductor material that is coated with a light-sensitive resist layer.

To facilitate understanding of the description, a Cartesian system coordinate system SKS is indicated, which reveals the respective orientation relationship of the components illustrated in the figures. The projection exposure apparatus WSC is of the scanner type. The x-axis extends perpendicularly to the plane of the drawing in figure 1. The y-axis extends towards the right. The z-axis extends downward. The object plane OS and the image plane IS both extend parallel to the x-y-plane. During the operation of the projection exposure apparatus, the mask and the substrate are moved synchronously or simultaneously in the y-direction (scanning direction) during a scan operation and are thereby scanned.

The apparatus is operated with the radiation from a primary radiation source RS. An illumination system ILL serves for receiving the radiation from the primary radiation source and for shaping illumination radiation directed onto the pattern. The projection lens PO serves for imaging the pattern onto the light-sensitive substrate.

The primary radiation source RS may be, inter alia, a laser plasma source or a gas discharge source or a synchrotron-based radiation source or a free electron laser (FEL). Such radiation sources generate a radiation RAD in the EUV range, in particular having wavelengths of between 5 nm and 15 nm. The illumination system and the projection lens are constructed with components that are reflective to EUV radiation in order that they can operate in this wavelength range.

The primary radiation source RS can be, in particular, a plasma source in which tin is excited to form a plasma with the aid of a carbon dioxide laser operating at a wavelength of 10.6 pm, which plasma emits, inter alia, the desired EUV radiation at approximately 13.5 nm with a relatively high intensity.

The primary radiation source RS is situated in a source module SM, which is separate from the illumination system ILL and also has, inter alia, a collector COL for collecting the primary EUV radiation. The source module SM generates during exposure operation a secondary radiation source SLS at a source position SP in an entrance plane ES of the illumination system ILL. The secondary radiation source SLS is the optical interface between the EUV radiation source or the source module SM and the illumination system ILL. The illumination system comprises a mixing unit MIX and a planar deflection mirror GM (also referred to as G mirror GM), which is operated under grazing incidence. The illumination system shapes the radiation and thereby illuminates an illumination field BF situated in the object plane OS of the projection lens PO or in the vicinity thereof. In this case, the shape and size of the illumination field determine the shape and size of the effectively used object field in the object plane OS. During operation of the apparatus, the reflective reticle is arranged in the region of the object plane OS. The plane OS is therefore also referred to as the reticle plane.

The mixing unit MIX substantially consists of two facet mirrors FAC1 , FAC2. The first facet mirror FAC1 is arranged in a plane of the illumination system which is optically conjugate with respect to the object plane OS. Therefore, it is also referred to as a field facet mirror. The second facet mirror FAC2 is arranged in a pupil plane of the illumination system that is optically conjugate with respect to a pupil plane of the projection lens. Therefore, it is also referred to as a pupil facet mirror.

With the aid of the pupil facet mirror FAC2 and the optical assembly disposed downstream in the beam path and comprising the deflection mirror GM operated with grazing incidence, the individual mirroring facets (individual mirrors) of the first facet mirror FAC1 are imaged into the illumination field.

The spatial (local) illumination intensity distribution at the field facet mirror FAC1 determines the local illumination intensity distribution in the illumination field. The spatial (local) illumination intensity distribution at the pupil facet mirror FAC2 determines the illumination angle intensity distribution in the illumination field OF.

The shape of the illumination field is determined substantially by the shape of the facets of the field facet mirror FAC1 , the images of which fall into the exit plane of the illumination system. The illumination field can be a rectangular field or else a curved field (ring field).

The beam-guiding region optically between the source position SP and the exit plane (plane of the image field) is the illumination beam path, in which the EUV radiation is successively incident during operation on the first facet mirror FAC1 , the second facet mirror FAC2 and the deflection mirror GM.

For further explanation, Figure 2 schematically illustrates a mirror arrangement SA, having a first facet mirror FAC1 and a second facet mirror FAC2. The first facet mirror FAC1 has a multiplicity of first facets F1 , which are in the form of elongate arcs in the exemplary embodiment shown. This shape of the first facets, however, should be understood to be merely exemplary. Only some of the facets are shown. The number of first facets in practice is typically significantly higher and can be more than 100 or even more than 300.

The second facet mirror FAC2 has a multiplicity of second facets F2, which are in the form of small dies in the exemplary embodiment shown, which again should be understood to be merely an example.

The first facets F1 are arranged on a first main body B1 of the first facet mirror FACT The first main body forms, together with the first facets it carries and any further components, e.g. attachment means, actuators etc., a first mirror module SM1.

The first mirror module SM1 can be installed in its entirety at the installation position that is provided therefor on an associated first carrier structure TS1 of the illumination system or else be demounted again in its entirety and removed. The orientation of the first mirror module SM1 in space, or relative to a reference coordinate system (e.g. the SKS of the housing of the illumination system), can be defined by way of the first module coordinate system MKS1.

The second facets F2 are analogously arranged on a second main body B2 of the second facet mirror, as a result of which a completely installable and replaceable second mirror module SM2 is formed. The orientation of the second mirror module SM2 in space, or relative to a reference coordinate system, can be defined by way of the second module coordinate system MKS2.

The relative orientation or position of a mirror module with respect to the associated carrier structures (frame structure of the illumination system) or to the system coordinate system that is linked thereto can be continuously variably or incrementally set with great accuracy in six degrees of freedom. Suitable adjustment means are provided herefor, which can also be referred to as tilt manipulators. Details of possible embodiments are described e.g. in DE 10 2016 203 990 A1.

Depicted in Fig. 2 by way of example are a few rays ST illustrating the EUV illumination beam path when the mirror arrangement is installed in an optical system and in operation. The rays ST start here from a first field plane FE1 (intermediate focus), and are then reflected by the facets F1 of the first facet mirror FAC1 onto the facets F2 of the second facet mirror FAC2. The rays are directed by the facets F2 of the second facet mirror FAC2 into a second field plane FE2, corresponding to the exit plane of the illumination system. Images IM of the facets of the first facet mirror FAC1 are thereby produced in the second field plane FE2, wherein, in more precise terms, the images of all first facets F1 are produced in the field plane FE2 by mutual superimposition. The superimposed images IM together form the illuminated illumination field BF.

Between the facets F1 of the first facet mirror FAC1 and the facets F2 of the second facet mirror FAC2 there is a unique assignment. That means that each facet F1 of the first facet mirror FAC1 is assigned a specific facet F2 of the second facet mirror FAC2. In Fig. 2, this is shown for a facet F1-A and a facet F1-B of the first facet mirror FAC1 and a facet F2-A and a facet F2-B of the second facet mirror FAC2. In other words, those rays ST which are reflected by the facet F1-A are incident exactly on the facet F2-A, and those used light rays which are reflected by the facet F1-B are incident on the facet F2-B etc. In this case, there is a 1 :1 assignment between the facets F1 of the first facet mirror FAC1 and the facets F2 of the second facet mirror FAC2.

In deviation from a 1 :1 assignment between the facets F1 and F2, it is also possible, however, that each facet F1 is assigned more than one facet of the facets F2. This is the case if the facets F1 are tiltable, that is to say can assume various tilted states, with the result that, in a first tilted state, each facet F1 is assigned a specific facet of the second facets F2, and, accordingly, in a different tilted state, is assigned a different facet of the second facets F2. Generally possible is a 1 :n assignment (n being a natural number) between the first facets F1 and the second facets F2, depending on how many states the first facets F1 can assume.

The illumination beam path is composed of many individual illumination channels, wherein an illumination channel extends in each case from the source position or from the intermediate focus FE1 , via a first facet F1 and a second facet F2 that is actually assigned to the first facet, into the illumination field.

In the exemplary embodiment shown of the mirror arrangement, the first facet mirror FAC1 is conjugate to the field plane FE2 and is therefore also referred to as a field facet mirror. In comparison, the second facet mirror FAC2 is conjugate to a pupil plane and is therefore also referred to as a pupil facet mirror.

In the case that the mirror arrangement is used in an illumination system of a projection exposure apparatus, the field plane FE2 is that plane in which the reticle is arranged, the pattern of which is intended to be imaged onto a wafer. In the case that the mirror arrangement SA is used in a mask inspection apparatus, the field plane FE2 is that plane in which the mask to be inspected is arranged.

In the exemplary embodiment in Fig. 1 , the illumination system comprises, in addition to a mirror arrangement having two facet mirrors FAC1 and FAC2, which acts as a mixing unit MIX, also the field-shaping mirror GM, which is operated under grazing incidence and is situated between the second facet mirror FAC2 and the exit plane or the object plane of the projection lens. This additional mirror can be favourable for reasons of installation space. In other exemplary embodiments, the illumination system has, aside from the two facet mirrors FAC1 and FAC2, no further mirrors in the illumination beam path.

In the illumination system ILL from Fig. 1 , all three mirror modules, i.e. the first field facet mirror FAC1 , the second field facet mirror FAC2 and the deflection mirror GM, are in each case replaceable in their entirety. That is to say, after releasing corresponding attachment means, they can be removed from their respective installation positions and be replaced by other components, for example components which are nominally of the same construction, without completely disassembling the illumination system.

After the mirror exchange, the illumination system should once again fulfil its desired function. In particular, the position of the illumination field in the exit plane should be situated sufficiently close to its desired position and the radiation should again be incident on the illumination field with the same angle distribution at a given illumination setting as before the mirror exchange. In order to ensure that the optical performance of the illumination system after exchange of a mirror module systematically corresponds again to the desired performance before the mirror exchange, auxiliary means are provided in the illumination system, which permit the systematic optimization of the mirror positions after installation such that the required optical performance can be achieved within an acceptable time period. The devices make possible a targeted adjustment of the illumination system at the site of its use, that is to say for example at the place of manufacture of semiconductor chips.

The illumination system is equipped with components of a measurement system MES which permits optical acquisition of information for determining the orientations of the mirror modules in the respective installation positions that are associated with the mirror modules, with the result that the adjustment can be systematically made on the basis of the measurement values which are obtained by the measurement system. The measurement system MES of the exemplary embodiment has the following components. A measurement light source module MSM contains a measurement light source MLS for emitting measurement light from the visible (VIS) spectral range. The measurement light source used can be, for example, a light-emitting diode (LED) or a laser diode. The measurement light source module MSM is arranged at the housing H of the illumination system outside the evacuable interior by way of first interface structures IF1 , can be mounted for measurement purposes and, if needed, removed again and may optionally be used for measurement purposes at a different location. The position of the measurement light source module with respect to the housing can be changed using positioning drives multiaxially both parallel to the central incidence direction and perpendicular thereto. One exemplary embodiment of a measurement light source module will be explained in greater detail in association with Fig. 5.

A switchable input coupling device IN is provided for coupling measurement light emitted by the measurement light source module MSM into the illumination beam path at an input coupling position upstream of the first facet mirror FACT The input coupling device comprises a plane mirror, which serves as the input coupling mirror MIN and which can be panned between a neutral position (illustrated in dashed lines) outside the illumination beam path and the input coupling position (illustrated in solid lines) using an electric drive. In the case of the example, the measurement light source module generates an image of the measurement light source MLS at the site of the source position SP (intermediate focus of the EUV radiation). The input coupling mirror MIN can be panned such that the measurement light beam is coupled into the illumination beam path at the site of the source position SP as if the measurement light source MLS were located at the site of the source position SP. With this arrangement, it is thus possible to imitate or simulate the source beam present in EUV operation by way of measurement light.

Situated behind the last mirror module of the illumination beam path, that is to say in the example of Fig. 1 downstream of the deflection mirror GM, in the region between the deflection mirror GM and the exit plane of the illumination system (object plane OS of the projection lens), is a switchable output coupling device OUT for coupling measurement light out of the illumination beam path, wherein the measurement light is coupled out after the measurement light has been reflected at each of the mirror modules of the illumination beam path. The switchable output coupling device comprises a plane mirror, which is used as the output coupling mirror MOUT and which can be panned between the neutral position (illustrated in dashed lines) outside the illumination beam path and the output coupling state (illustrated in solid lines) using an electric drive. In the output coupling state, the output coupling mirror reflects the measurement light coming from the deflection mirror GM in the direction of a detector module position, in which a detector module DET is arranged. The detector module DET is attached to the outside of the housing H of the illumination system in its detector position using second interface structures IF2 and can be adjusted in terms of its position via electrically actuable positioning drives. One exemplary embodiment of a detector that can be used for this purpose is described in DE 10 2016 203 990 A1. The disclosure of this document in this respect is incorporated by reference in the content of this description.

All controllable components of the measurement system MES are connected in signal- transmitting fashion to the control unit SE of the measurement system in the ready-for-operation installed state of the measurement system. Also situated in the control unit is an evaluation unit for evaluating the measurement values obtained using the measurement light, which measurement values represent the adjustment state of the mirror modules within the illumination system.

At least one mirror surface of one of the mirror modules is provided with an IR diffraction structure for diffracting radiation having a wavelength in the infrared range. In the case of the example, the diffraction structure is intended to be suitable to diffract disturbing infrared light, in particular having a wavelength of approximately 10.6 pm, out of the illumination beam path or the used beam path. The IR diffraction structure can be embodied e.g. as a binary phase grating.

Analogously to one example from DE 10 2012 010 093 A1 , an IR diffraction structure DS-IR, e.g. in the form of a binary phase grating, can be embodied for example at each mirror surface of the individual facets of the first facet mirror FAC1. Fig. 1 illustrates such a grating-like structure for one of the facets schematically in an enlarged manner. The binary phase grating preferably has a grating period p dimensioned such that IR radiation which is to be masked out and which is incident on a facet of the first facet mirror FAC1 is not incident on the second facet - assigned to said facet - of the second facet mirror FAC2 and is thus diffracted out of the illumination channel formed by the two facets. The binary phase grating can have for example a grating period p which is adapted to the embodiment of the facets of the second facet mirror FAC2 in such a way that the 1 st and -1 st orders of diffraction of I R radiation of the wavelength to be masked out are imaged onto facets of the second facet mirror which lie adjacent to the facet on which the image of an imaged radiation becomes located. The pupil facets onto which the 1 st and -1 st orders of diffraction of the IR radiation to be masked out are imaged are preferably aligned e.g. by tilting in such a way that the radiation to be masked out is not imaged into the illumination field BF to be illuminated in the object plane OS. The radiation to be masked out can be deflected to a light trap, for example. Figs. 3A and 3B show two examples of the possible embodiment of IR diffraction structures DS- IR in the form of binary phase gratings. The figures each show a sectional view through a multilayer structure of an EUV mirror perpendicular to the reflective mirror surface. The figures reveal the successive layers - extending parallel to one another - of high refractive index and, relative thereto, low refractive index material for forming the EUV-reflecting multilayer. The macroscopically structured mirror surface has in each case front regions V and rear regions H, which are aligned parallel to one another and are offset relative to one another by a predefined offset d in the direction of their surface normals. In the case of the example, the front regions V and the rear regions H have identical widths in a direction perpendicular to their surface normals. The width of the front regions V is also referred to as ridge width, and the width of the rear regions H is also referred to as furrow width. Usually at least one front region of this type and one rear region of this type are provided at a mirror surface. However, a multiplicity of regions of this type can also be provided, in particular in such a way that the regions embody a grating structure having a grating period p, wherein the grating period is also referred to as grating constant or pitch. In particular, a binary diffraction grating can be involved.

The grating period p can generally be chosen in such a way that electromagnetic radiation having wavelengths above that EUV radiation which is used during operation of the illumination system is diffracted away. In the case of the example seeking to mask out IR radiation, the grating period can be for example in the range of 780 nm to 1 mm, in particular in the micrometres range, e.g. in the range around 10.6 pm.

As shown in Fig. 3A, the transitions between front regions V and rear regions H can be steep in a steplike manner. It is also possible for the ridges and furrows not to have perpendicular sidewalls, but rather oblique sidewalls, as shown by way of example in Fig. 3B. This may be preferred for reasons of better producibility, for example.

Fig. 4 schematically illustrates components of an illumination system ILL in accordance with another exemplary embodiment. For reasons of clarity, the same reference signs as in the case of the exemplary embodiment in Fig. 1 are used for structurally and/or functionally identical or similar components.

The primary radiation source RS can be a plasma source, for example, in which tin is excited to form a plasma with the aid of a carbon dioxide laser operating at a wavelength of 10.6 pm, said plasma emitting, inter alia, the desired EUV radiation at approximately 13.5 nm with high intensity. The radiation is focused by a collector COL. A suitable collector is known for example from EP 1 225 481 A1. Downstream of the collector, the EUV radiation propagates through an intermediate focal plane ES to the downstream components of the illumination system ILL. The source position SP is situated in the intermediate focal plane ES. The radiation coming from the source position firstly propagates onto the first facet mirror FAC1 (field facet mirror) and is reflected by the illuminated facets thereof in the direction of the facets of the second facet mirror FAC2. The first field facet mirror FAC1 is arranged in a plane of the illumination system which is optically conjugate with respect to the illumination plane OS. Since the reticle to be exposed (the mask) is arranged in this plane, this plane is also referred to as reticle plane. The pupil facet mirror FAC2 is arranged in a pupil plane of the illumination system that is optically conjugate with respect to a pupil plane of the downstream projection optical unit. With the aid of the pupil facet mirror FAC2 and a downstream transfer optical unit TO having an imaging effect, illuminated individual facets of the first facet mirror FAC1 (field facets) are imaged into the illumination field such that they are at least partly superimposed there.

The transfer optical unit TO of the example from Fig. 4 contains only a single optical component in the form of a condenser CO having a generally concave mirror surface, on which all individual channels of the illumination beam path are incident. In other exemplary embodiments (for example in a manner similar to the illumination system from Fig. 1 from DE 10 2012 010 093 A1 ), the imaging transfer optical unit can also contain more than one component, for example two or three reflective components disposed one after another.

In the example in Fig. 4, the concave mirror surface of the condenser CO is configured with an IR diffraction structure DS-IR, which has an effect such that at least one portion of incident IR radiation is diffracted out of the illumination beam path in such a way that it does not pass into the illumination field BF.

In the case of the example, the IR diffraction grating at the condenser CO is embodied as a reflective phase grating (echelon grating) and has a grating depth d of one quarter of the wavelength of the infrared radiation to be suppressed, such that the 0 order of diffraction for the wavelength of the IR light is suppressed. The grating period (pitch) is chosen such that the diffraction angle for the first orders of diffraction for the wavelength of the IR light is approximately of the order of magnitude of 10 mrad. A sufficiently great separation of the higher orders of diffraction of the IR light from the illumination beam path or used beam path thus arises in the reticle plane (object plane OS of the projection lens (not illustrated)).

The phase grating preferably has a grating depth d or furrow depth corresponding to precisely one quarter of a wavelength to be masked out of IR radiation. The grating depth or furrow depth d is in particular in the range of 2 pm to 3 pm, e.g. in the range of 2.5 pm to 2.7 pm, preferably approximately 2.65 pm. The diffraction structure has a grating period p (pitch p) of at most 5 mm, in particular at most 3 mm, in particular at most 2 mm, in particular at most 1 mm. In this case, a smaller grating period leads to a larger deflection angle for the first order of diffraction of the IR radiation. The deflection angle can be e.g. at least 3 mrad, in particular at least 5 mrad.

In Fig. 4 the central beam EUV converging in the direction of the illumination field symbolizes the illumination beam path of the EUV radiation (desired during operation). Illustrated next to that on the left and right are beams which, proceeding from the mirror surface of the collector CO, do not converge in the direction of the illumination field BF, but rather are incident in the object plane OS outside the illumination field BF. These beams represent the +1 st order (+1 IR) and the -1 st order (-1 IR) of the infrared radiation diffracted by the IR diffraction structure.

The schematic intensity diagram directly above the object plane in Fig. 4 shows the spatial intensity distribution of the IR radiation (HR) in the object plane OS with intensity maxima at the location of the incidence of the first orders of diffraction and largely suppressed IR intensity in the region of the illumination field BF.

For the measurement in association with the geometric adjustment of the illumination system (during the first adjustment during the original production or during an adjustment for re- establishing the performance, e.g. after mirror exchange), visible light, for example having a wavelength of the order of magnitude of approximately 500 nm, passes through the entire illumination beam path. The measurement light is provided by the measurement light source module MSM. It can be coupled in, coupled out and detected as described in association with Fig. 1. Some possible measurement methods are described in DE 10 2016 203 990 A1. They can be used here. The disclosure of said document is incorporated by reference in the content of this description.

Said measurement light is also incident on the IR diffraction structures. These structures do not have a significant diffractive effect for the EUV light used during productive operation. The diffraction angles of the higher orders are in the range of a few prad and are thus practically invisible in the reticle plane. For the light from the visible spectral range that is used during the measurement, that is to say the measurement light, the diffraction angles are approximately of the order of magnitude of 0.5 mrad, however, with the result that on the reticle plane the -1st and the +1 st orders of diffraction of the measurement light are situated at a distance of approximately 1 mm from the zero order of diffraction. It has been found that the measurements carried out with the aid of the measurement system can thereby be disturbed. In particular, the position measurement of the field position and the measurement of pupil spots can be adversely affected.

In order to provide a remedy here, for the measurement the wavelength of the visible light is chosen substantially such that the higher orders of diffraction arising at the IR diffraction structure DS-IR are largely or completely suppressed. With normal incidence of the measurement light on a diffraction grating, this condition is satisfied if the grating depth d corresponds precisely to a half-integer multiple of the wavelength l of the measurement light, that is to say d= h·l/2. For the case of oblique incidence of the measurement light on a diffraction grating with an angle of incidence a with respect to the surface normal, it correspondingly holds true that d/cos(a)= h·l/2 where n= 1 , 2, 3.

The schematic intensity diagram directly above the diagram IIR(x) in Fig. 4 shows the spatial intensity distribution of the measurement light from the visible spectral range (IVIS) in the object plane OS. The majority of the intensity is found in the 0 order of diffraction within the illumination field; higher orders are largely suppressed.

In the illumination system of the type illustrated here, the different illumination channels (individual channels of the beam path from the source position via a field facet and a pupil facet and, if appropriate, one or more mirror surfaces from the transfer optical unit to the illumination field) have, in principle, different angles of incidence on the IR diffraction structure. For sufficiently great suppression of the higher orders during measurements with visible light, it can be advantageous to select or to set the wavelength for the measurement separately for each illumination channel or for groups of illumination channels with relatively similar angles of incidence.

For the exemplary illustration of quantitative relationships, Fig. 5 shows a d-l diagram concerning the ratio between the step depth d of a binary phase grating designed for suppressing 10.6 m radiation and the wavelength l of the measurement light for different angles of incidence a (a=0, a=10°, a=20°) of measurement light from the VIS range between 450 nm and 600 nm. It is evident that measurement light having a wavelength tending to be greater should be chosen if the measurement light is incident on the IR diffraction structure at relatively greater (average) angles of incidence.

For the channel-dependent setting of suitable angles of incidence, the used measurement system for generating the measurement light can comprise a measurement light source module that is tunable continuously variably or in steps, such that the measurement light source module can generate different required wavelengths of measurement light and furthermore allows the combination of different beam angles with suitable wavelengths. One exemplary embodiment is explained in association with Fig. 6.

It has proved worthwhile in many cases to use a light-emitting diode (LED) as primary light source for system metrology with visible light. Taking account of typical angle-of-incidence spectra in the illumination system and the above-described relationships between the grating dimensions of a grating that is diffractive for IR radiation and the wavelengths in the visible wavelength range that are used for the measurement, it emerges that the wavelength range required to achieve the sought suppression of the higher orders of diffraction for all possible angles of incidence is approximately 20 nm to 30 nm wide. Expediently, the measurement light having a bandwidth AA<2nm should moreover be sufficiently narrowband in order, with a correctly set wavelength, to be able to suppress the greatest possible proportion of the measurement light in the higher orders of diffraction.

Fig. 6 shows components of a measurement light source module MSM which can satisfy these parameters. In this case, a primary radiation source MLS that is sufficiently broadband in the required wavelength range is used, for example a light-emitting diode (LED) or a thermal radiator. Said primary radiation source is firstly imaged into infinity with the aid of a first Fourier lens element FL1 in order to generate parallel beams. For this purpose, the distance between the first Fourier lens element FL1 and the light source corresponds precisely to the lens element focal length. Situated in the parallel beam path downstream of the first Fourier lens element is a rotatably mounted interference filter IF exhibiting sufficiently narrowband transmission. Said filter can be embodied for example as a dielectric mirror or the like. The transmitted wavelength or the spectral position of a narrow transmitted wavelength range can be set within certain limits by way of the tilt angle b of the filter (measured for example between the optical axis OA of the set-up and the normal to the surface of the interference filter). A second Fourier lens element FL2 images the primary light source into an exit plane AE that is optically conjugate with respect to the plane of the primary light source. In the back focal plane of the second Fourier lens element FL2, there thus arises an image of the primary light source or a secondary light source SMLS only with the transmitted wavelengths. If the entire required wavelength range cannot be achieved by way of a single interference filter, a plurality of exchangeable interference filters can be provided, if necessary, which can optionally be moved into the beam path manually or, if appropriate, in an automated manner (for example by way of a turret arrangement or a linear stage). The measurement light source module is furthermore designed such that different beam angles of the emitted measurement light can be emitted. In this case, a specific wavelength of the measurement light can be set as necessary for each beam angle or a group of similar beam angles. This functionality can be referred to as channel-dependent wavelength adaptation. A pupil plane PE is situated between the entrance plane, in which the measurement light source MLS is situated, and the exit plane AE, said pupil plane being a plane that is Fourier- transformed with respect to the entrance plane and the exit plane. Located in the region of the pupil plane is a stop CS having an aperture MO, through which a selected portion of the measurement light can pass. The position of the aperture MO is freely selectable in two dimensions within the pupil plane. The displaceable stop CS can thus be used to select a specific portion of the measurement light for emission. The location of the through-opening in the pupil plane here determines the angle of incidence of the measurement light, which was allowed to pass (dashed line), at the location of the secondary measurement light source SMLS and consequently also the emission angle of the measurement light from the measurement light source module. In this way, various individual channels or channel groups of the illumination system can be selected for a measurement.

The arrangement can be implemented so as to give rise to the image of the primary measurement light source MLS, that is to say the secondary measurement light source SMLS at the source position SP in the entrance plane IS of the illumination system.

Other configurations are likewise possible. As illustrated in Fig. 7, it is possible, for example, to couple the measurement light downstream of the second Fourier lens element FL2 into an optical glass fibre or an optical glass fibre bundle or some other optical waveguide LL and to couple it out again at the source position SP. In this case, the entire optical set-up of the measurement light source module MSM can be arranged spatially separately from the illumination system ILL and merely the light exit of the optical waveguide serves as an effective light source at the illumination system. That can be useful for example for a measurement in the field (in the case of an illumination system installed in the projection exposure apparatus at the location of use on the part of the end customer) for example in the event of a mirror exchange. Here a simple channel selection (selection of the angle of incidence) in a manner similar to Fig. 6 is not possible, since the measurement light passes through an optical fibre and, downstream thereof, once again the complete emission angle range of the fibre is illuminated.

In principle, a tunable laser (for example an external cavity diode laser) can also be used as a light source. When a laser is used as a light source, additional elements for destroying coherence and for adapting the optical properties to the input parameters of the illumination system should be present, for example a rotating diffusing plate for destroying coherence.